How to Install Jupyter Notebook for Machine Learning: Step-by-Step Guide (Updated August 2026)
Trust me — the installation step is where more beginners give up than at any other point in their machine learning journey. You Google install Jupyter, get five different answers, try one, get an error, and close the browser. I have seen this with students from Pune, Sambhajinagar, and Sangli who eventually walked into our center saying they could not even get started. The good news is it takes under 10 minutes if you do it in the right order. NASSCOM-Deloitte estimates India needs 1.25 million AI and ML professionals by 2027 — every student who sets up a working ML environment is already ahead of the curve. This guide covers the exact steps, common errors, and the workflow you need once Jupyter is running.
- Install Python first from python.org — tick "Add Python to PATH" during setup or pip will not work.
- Open CMD, type pip install jupyter, press Enter. Wait 1–3 minutes for download.
- Navigate to your project folder in CMD, then type jupyter notebook — your browser opens the Dashboard.
- Create notebook: New → Python 3. Run code: Shift+Enter. Save: File → Save As with a clear name.
Prerequisites: What You Need Before Installing Jupyter
You need Python installed on your machine — that is the only prerequisite. Jupyter runs on Python, and pip (Python package manager) handles the rest. Download Python from python.org — version 3.9 or above is recommended as of 2026. During installation on Windows, tick the checkbox that says Add Python to PATH. This is the step most beginners miss, and it causes the dreaded pip is not recognized error in CMD later. Verify: open CMD and type python --version. You should see something like Python 3.11.4. If you do not, Python is not in your PATH — reinstall with the PATH checkbox ticked. On Mac and Linux, Python is often pre-installed but may be version 2.x — install Python 3.x explicitly.
Installing Jupyter Notebook: The Exact Steps
Step 1: Open CMD on Windows or Terminal on Mac or Linux. Step 2: Type pip install jupyter and press Enter. pip downloads and installs Jupyter and all its dependencies — you will see a progress bar. Internet is required for this step. Step 3: Wait for completion — on a typical broadband connection, this takes 1 to 3 minutes. Step 4: Verify by typing jupyter --version. You should see version numbers for core Jupyter components. Jupyter is now installed. For JupyterLab — the newer, more powerful interface with multi-tab support, built-in terminal, and file browser — use pip install jupyterlab and launch with jupyter lab. For most beginners, classic Jupyter Notebook is the right starting point.
How to Open and Use Jupyter Notebook for the First Time
Launching matters. Do not just type jupyter notebook from any directory. First navigate to the folder where you want to store your notebooks. In CMD: type your drive letter (C:), then cd Documents\ML_Projects to navigate to your project folder. Then type jupyter notebook. Your browser opens the Dashboard — a file manager showing all files in that folder. To create a new notebook: click New then Python 3. The cursor is inside the first cell. Type print("Machine Learning starts here") and press Shift+Enter. Output appears directly below the cell. A new empty cell appears below — continue writing. To save: File then Save Notebook As, then type a clear name like Session01_Intro and press Enter. The file saves as .ipynb in your project folder. Key shortcuts: DD to delete a cell, B to insert below, A to insert above, M to change a cell to markdown mode.
Understanding the Jupyter Workflow for ML Projects
The real power of Jupyter comes from its structured workflow for ML projects. Cell 1: Import libraries (pandas, numpy, matplotlib, sklearn). Cell 2: Load data with pd.read_csv("dataset.csv"). Cell 3: Explore with df.describe(), df.info(), df.head(). Cell 4: Clean data — handle nulls, remove duplicates. Cell 5: Visualize with matplotlib charts and seaborn heatmaps. Cell 6: Feature engineering — create or transform features. Cell 7: Split data with train_test_split. Cell 8: Train model with clf.fit(X_train, y_train). Cell 9: Evaluate with accuracy_score and confusion_matrix. Cell 10: Interpret with feature importance or SHAP. The cell-by-cell approach means you re-run Cell 8 with a different model without re-running cells 1 through 7. That is the workflow that makes Jupyter the industry standard for ML research at companies like TCS, Infosys, and data teams across Maharashtra.
Common Jupyter Installation Errors and How to Fix Them
pip is not recognized: Python is not in your PATH. Reinstall Python with Add to PATH checked. jupyter is not recognized after install: the Scripts folder is not in PATH. Find Python installation (usually C:\Users\YourName\AppData\Local\Programs\Python\PythonXX\Scripts), add it to system environment variables, restart CMD. Browser does not open automatically: type http://localhost:8888 manually in your browser — Jupyter is running, it just did not launch the browser automatically. Address already in use: another Jupyter instance is running. Use jupyter notebook --port 8889 to run on a different port. Kernel not found or dead kernel: restart the kernel from the Kernel menu. If persistent, run pip install ipykernel then python -m ipykernel install --user. Slow startup: normal on first launch — Jupyter builds its configuration files. Subsequent launches are much faster.
| Step | Command / Action | What Happens |
|---|---|---|
| Install Python | Download from python.org (tick Add to PATH) | Python and pip available in CMD |
| Install Jupyter | pip install jupyter | Jupyter and all dependencies installed |
| Launch | jupyter notebook (from project folder) | Browser opens Dashboard |
| Create Notebook | Click New → Python 3 | Empty notebook opens in new tab |
| Run Code | Shift+Enter | Output appears below the cell |
| Save | File → Save Notebook As → type name → Enter | .ipynb file saved to project folder |
Get the Machine Learning Brochure + Fees + Batch Dates on WhatsApp
Free 1:1 counselling. Placement track record. CMYKPY/PMKVY eligibility check.
💬 Get Brochure on WhatsApp📞 Call 7039169629About the author: Amit Kulkarni. 8 yrs leading IT training at ABC Trainings, ex-Infosys. Trained 2000+ students in Python, Machine Learning and Data Science across Pune, Sambhajinagar and Sangli..
Visit Our Centers
- Wagholi (Pune): 1st Floor, Laxmi Datta Arcade, Pune-Ahilyanagar Highway. Call 7039169629
- Hadapsar (Pune HQ): 1st Floor, Shree Tower, opp. Vaibhav Theater, Magarpatta. Call 7039169629
- Cidco (Chh. Sambhajinagar): Kalpana Plaza, opp. Eiffel Tower, N-1 Cidco. Call 7039169629
- Osmanpura (Chh. Sambhajinagar): S.S.C Board to Peer Bazar Road, near Jama Masjid. Call 7039169629
- Sangli: Shubham Emphoria, 1st Floor, Above US Polo Assn., Sangli-Miraj Rd, Vishrambag. Weekend batches available. Call 7039169629
FAQs
Do I need to install Jupyter separately if I already have Anaconda?
No. Anaconda includes Jupyter Notebook by default — open Anaconda Navigator and launch Jupyter from there. Anaconda also installs most ML libraries (pandas, numpy, scikit-learn, matplotlib) in one step. For beginners who do not want to manage packages manually, Anaconda is the easier route. The pip install route works fine too if you prefer a lighter setup.
Can I use Jupyter Notebook on a low-specification laptop?
Yes, for learning purposes. Jupyter itself is lightweight — resource usage depends on what your code does. Basic ML with scikit-learn on datasets up to 100,000 rows runs fine on 4GB RAM. For deep learning with TensorFlow or PyTorch, 8GB RAM is the minimum and GPU access via Colab is recommended. An i3 processor with 4GB RAM is enough to complete a full ML course using Jupyter Notebook.
How do I install ML libraries like pandas and scikit-learn for use in Jupyter?
Use pip from CMD: pip install pandas numpy scikit-learn matplotlib seaborn. After installation, these are immediately available inside Jupyter — just import pandas as pd at the top of your notebook. If you get a module not found error inside Jupyter after pip install, it means pip installed to a different Python environment. Use pip install --user pandas or switch to Anaconda.
Is Google Colab better than local Jupyter Notebook for machine learning beginners?
For beginners with a slow laptop or limited data, local Jupyter is more practical — no internet needed once installed. For beginners with good internet, Colab is excellent: no installation, free GPU access, and Google Drive integration. Many students use both — Colab for GPU-heavy deep learning tasks, local Jupyter for regular data analysis and ML coursework.
